AI use cases and applicationsFull notesSummaryRingkasanStoriesPracticeBusiness Metrics for Generative AI6 exam-style questions on this lesson.Question 1 of 6A company launches a generative AI customer support assistant. Which metric best measures its success?ACustomer satisfactionBConversion rateCAverage revenue per userDNumber of GPUs usedCheck answerQuestion 2 of 6An e-commerce site adds AI-generated product recommendations to increase upselling. Which metric tracks revenue earned per user?AEfficiencyBAverage revenue per user (ARPU)CCross-domain performanceDPerplexityCheck answerQuestion 3 of 6A company uses generative AI to write marketing emails and wants to know what share of recipients go on to make a purchase. Which metric fits?AEfficiencyBUser satisfactionCConversion rateDROUGECheck answerQuestion 4 of 6An enterprise assistant is used by HR, finance, and legal teams. Which metric shows how well it handles tasks across these different areas?AConversion rateBARPUCBLEUDCross-domain performanceCheck answerQuestion 5 of 6A company automates document drafting with generative AI and wants to measure time and cost saved. Which metric fits?AEfficiencyBConversion rateCARPUDUser satisfactionCheck answerQuestion 6 of 6Why isn't model accuracy enough to judge a generative AI application?AAccuracy can't be measured for generative AI outputBBusiness metrics are needed to show outcomes and ROICRegulators forbid using accuracy metrics for AIDGenerative AI models are always 100% accurateCheck answerModels7 exam-style questions on this lesson.Responsible AI6 exam-style questions on this lesson.